Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Apple Silicon M1 Die-Shots

techinsights.com

Post reply on HN
Print view View original

101–110 of 114 posts

Re: Apple Silicon M1 Die-Shots

#101

Post by trollied »

joakleaf wrote: post #73

I wonder what they'll do for "M1X" and pro-machines? Seems like heat isn't an issue, so can they increase the clock-speed of the M1? And will they create a larger die with 2xGPU and 2xCPUs, or will they "stitch" multiple M1 dies together on one package to have a higher yield and reduce supply chain concerns? An M1X with 2xGPU will bring it to Radeon 5500M level which is what I have in my current 16", but still not qu…

I'd like to see a pro machine with daughterboard slots, allowing you to add extra CPUs. A bit like the Acorn RISC PC. I guess Apple have already done this to seom extent with the FPGA cards that you can buy for the current Mac Pro. I do wonder if that was a taste of things to come?

Re: Apple Silicon M1 Die-Shots

#102

Post by hombre_fatal »

JohnJamesRambo wrote: post #70

Can we step back and marvel at the absurdity that one of the most powerful CPUs ever created was for a phone?

It's not absurd at all, nor surprising that the most power efficient chip was compelled by the most portable general computing device. Plugged-in desktop computers certainly weren't going to bring us here.

This just sounds like "DAE think phone=bad XD" anti-intellectualism.

Re: Apple Silicon M1 Die-Shots

#103

Post by faitswulff »

JohnJamesRambo wrote: post #70

Can we step back and marvel at the absurdity that one of the most powerful CPUs ever created was for a phone?

If you were used to using mainframes, you could have said much the same thing about desktop CPUs in the 90s. "That tiny thing doesn't even fill up a room!"

Re: Apple Silicon M1 Die-Shots

#104

Post by valuearb »

joakleaf wrote: post #92
Hamuko wrote: post #89

Earlier quoted context omitted.

Well, the prevailing rumor currently is a processor with 8 high-performance cores (as opposed to the 4 on the M1). I don't know if the amount of high-efficiency cores would also be doubled so that it's like 2xCPU when compared to the M1.

Yes, that's the M1X rumor, but that isn't going to be enough to compete with high-end CPUs in iMac and Mac Pros. For those, we have the 32-core rumor (but that is a massive die). And doubling the M1-GPU will not be enough to even surpass current Macbook Pro offerings. So they'll also need to do something more on that front.

For the truly high end they will need an external GPU. Removing the integrated GPU should get up enough space for lots more Firestorm cores.

And given that Firestorm is the fastest available CPU core, they probably don’t need 32 of them to match the highest multicore performance levels of existing AMD and Intel CPUs.

Re: Apple Silicon M1 Die-Shots

#105

Post by jeffbee »

dan1234 wrote: post #63
jeffbee wrote: post #56

Earlier quoted context omitted.

The benefit is they can charge you $200 for memory they bought from Samsung for $15. There’s no performance advantage to the user.

I’d love to know where you’re picking up LPDDR4X-4266 RAM for $15?

Are you actually asking? The spot price for LPDDR4X 16G is only $9. I was actually being generous. No doubt the contract price is far lower.

Re: Apple Silicon M1 Die-Shots

#106

Post by ksec »

dismalpedigree wrote: post #85

I work at TechInsights. Depending on what the community is looking to know, there may be some things we can share publicly. I’m in the software group and not intimately involved with the RE parts, but I will act as an ambassador and see if I can get the right people involved. Throw some comments on this thread and we’ll see what we can do to help. Might be next week due to holidays.

I am actually interested in the Non-Labeled part which no one on the Internet have even guessed what they are. Image Signal Processors ( ISP ), Video Decoder and Encoder, there should be a FPGA somewhere, Secure Enclave, NAND Controller, PCI-Express Controller / USB3 / Thunderbolt etc.

Mostly because all the other part are already well known.

Re: Apple Silicon M1 Die-Shots

#107

Post by ksec »

joakleaf wrote: post #73

I wonder what they'll do for "M1X" and pro-machines? Seems like heat isn't an issue, so can they increase the clock-speed of the M1? And will they create a larger die with 2xGPU and 2xCPUs, or will they "stitch" multiple M1 dies together on one package to have a higher yield and reduce supply chain concerns? An M1X with 2xGPU will bring it to Radeon 5500M level which is what I have in my current 16", but still not qu…

>Seems like heat isn't an issue, so can they increase the clock-speed of the M1?

They are running into their current node limits, so dont expect a 4Ghz+ coming unless they decide to remake the whole thing in High Performance Node. ( Which means it will cost a lot more for a relatively niche volume.)

>I wonder if they can make a 4xGPU

GPU increase should be given, and die size / yield would not be a concern / problem. The question is how will they handle the additional memory bandwidth requirement? As you can see in the Die-Shots those DDR I/O side are already take up much of the side space. Going with HBM2? Or LPDDR5?

>32xcore M1-family CPU, but I imagine the yield of such a beast would be horrible,

A 32x CPU / GPU Core M1 will only be 4x increase from its current form. If you exclude I/O part, the CPU + GPU + Cache is only a little more than 50% of die size, or ~60mm2. 4x That is only 240mm2+. The final product would be a sub 350mm2 if you include all the additional. Which is roughly inline with the current estimate of AWS Graviton 2 die size ( ~400mm2).

But yes, M1 ( A14X ) isn't really that interesting. The most exciting part is how they will go about doing high performance chip. Especially on iMac and Mac Pro.

Re: Apple Silicon M1 Die-Shots

#108

Post by Someone »

murukesh_s wrote: post #98
jclardy wrote: post #95

Earlier quoted context omitted.

CPU speeds have been similar to competitors until around the A10 era a few years ago. The real differentiator has always been software + hardware integration happening inside the same company. The hardware team gets the insight of the software team on what areas of performance should be looked at, and the software team can optimize for the exact specifications of a future CPU line.

i thought Apple crushed other chipset in single core and managed to stay ahead in overall. https://www.phonearena.com/news/apple-a14-vs-a13-vs-snapdrag... .

You’re saying The iPhone is mostly popular because the A13/14 are faster than the competition.

The grandfather points out that iPhones up to the A10 had comparable performance as the competition, and still were popular, so having faster CPUs can’t be the major reason for that popularity.

Re: Apple Silicon M1 Die-Shots

#109

Post by Someone »

viktorcode wrote: post #62
userbinator wrote: post #38

Earlier quoted context omitted.

The degree to which they hype UMA (a term of art within the PC industry for decades ) on that page is amusing --- if Intel did the same sort of marketing for its integrated GPUs, they would almost certainly be laughed at.

From what I've read it seems to be different from Intel's UMA. It's cache-friendly single pool of memory directly accessible from different cores of the SoC: CPU, GPU, NPU etc. Since there's no division of RAM between different parts of the SoC, copying operations aren't needed.

I also read somewhere that they made sure the data formats used by the various modules are identical. Without that, you have to copy data to do format conversions. I don’t know how much of a problem that is in general, though.

Re: Apple Silicon M1 Die-Shots

#110

Post by dooferlad »

joakleaf wrote: post #73

I wonder what they'll do for "M1X" and pro-machines? Seems like heat isn't an issue, so can they increase the clock-speed of the M1? And will they create a larger die with 2xGPU and 2xCPUs, or will they "stitch" multiple M1 dies together on one package to have a higher yield and reduce supply chain concerns? An M1X with 2xGPU will bring it to Radeon 5500M level which is what I have in my current 16", but still not qu…

(microelectronics graduate in 2001, turned software engineer, I have worked for ARM and Imagination Technologies on simulations of CPUs, GPUs, networking products and other bits of SoCs)

Yield wouldn't be that bad - ARM CPUs are small in comparison to their x86 cousins. I would expect that you would get to 32 CPUs the same way everyone else does - build a 32 core design and blow enable fuses on bad CPUs, then badge as appropriate.

To get a desktop class GPU they will need to address memory bandwidth. I am assuming they are using a tile based rendering process so memory bandwidth is less of an issue in comparison to scan line rendering (do desktop GPUs still do that?) I would assume that they are enjoying the benefits of the system level cache with the CPU and GPU sharing. I would expect there to be some careful benchmarking of increasing GPU cores, memory speed, memory size and system cache size going on at Apple.

There isn't anything stopping Apple from supporting external GPUs, but it would require a new SoC with more PCIe lanes. External buses are much more power hungry and take up space on the die. I don't have a mental model of how much space plonking a PCIe x16 interconnect would cost in terms of area or power (taking into account you need to get that data across the chip too), but my gut reaction is that it isn't likely unless there is a customer use case they can't solve on chip.

Post reply on HN